Search

The anisotropic diffusion coefficients of the radioactive tracers cadmium ($115m$) and gold (198) in single crystals of high-purity zinc, in directions parallel and perpendicular to the hexagonal axis, were measured by the standard sectioning technique. Cadmium diffused at a rate faster than that of self-diffusion and with ${D}_{\ensuremath{\perp}}g{D}_{\mathrm{II}}$. Gold diffused at a rate slower than that of self-diffusion and with ${D}_{\mathrm{II}}g{D}_{\ensuremath{\perp}}$. The results of the measurements are as follows: For cadimum ($115m$), ${D}_{\mathrm{II}}=(0.11\ifmmode\pm\else\textpm\fi{}0.01)\mathrm{exp}[\ensuremath{-}\frac{(20.54\ifmmode\pm\else\textpm\fi{}0.09)\ifmmode\times\else\texttimes\fi{}{10}^{3}}{\mathrm{RT}}]$ ${\mathrm{cm}}^{2}$/sec, ${D}_{\ensuremath{\perp}}=(0.12\ifmmode\pm\else\textpm\fi{}0.01)\mathrm{exp}[\ensuremath{-}\frac{(20.42\ifmmode\pm\else\textpm\fi{}0.03)\ifmmode\times\else\texttimes\fi{}{10}^{3}}{\mathrm{RT}}]$ ${\mathrm{cm}}^{2}$/sec; for gold (198), ${D}_{\mathrm{II}}=(0.97\ifmmode\pm\else\textpm\fi{}0.22)\ifmmode\times\else\texttimes\fi{}\mathrm{exp}[\ensuremath{-}\frac{(29.73\ifmmode\pm\else\textpm\fi{}0.26)\ifmmode\times\else\texttimes\fi{}{10}^{3}}{\mathrm{RT}}]$ ${\mathrm{cm}}^{2}$/sec, ${D}_{\ensuremath{\perp}}=(0.29\ifmmode\pm\else\textpm\fi{}0.12)\mathrm{exp}[\ensuremath{-}\frac{(29.72\ifmmode\pm\else\textpm\fi{}0.45)\ifmmode\times\else\texttimes\fi{}{10}^{3}}{\mathrm{RT}}]$ ${\mathrm{cm}}^{2}$/sec. On the basis of the vacancy mechanism one can interpret the fast diffusion of cadmium in terms of an appreciable binding between the vacancy and the diffusing ion; conversely the slow diffusion of gold in terms of a repulsion between the vacancy and the diffusing ion.

The self-diffusion of gold was measured in single crystalline specimens of 99.8 and 99.999% purity in the temperature range from 600-954\ifmmode^\circ\else\textdegree\fi{}C, by using ${\mathrm{Au}}^{198}$ as tracer and applying the sectioning technique. The temperature dependence of the self-diffusion coefficient is given by $D=(0.031\ifmmode\pm\else\textpm\fi{}0.004)\mathrm{exp}[\ensuremath{-}\frac{(39360\ifmmode\pm\else\textpm\fi{}280)}{\mathrm{RT}}] {\mathrm{cm}}^{2}/sec.$ Together with recently reported data on self-diffusion of copper, silver, and lead, this result is compared with empirical and theoretical predictions.

The sectioning technique was used to study the diffusion of radioactive tracers in high-purity zinc single crystals. Diffusion both parallel and perpendicular to the hexagonal axis was measured. The diffusion of ${\mathrm{Cu}}^{64}$ was measured over the temperature range from about 338\ifmmode^\circ\else\textdegree\fi{}C to 415\ifmmode^\circ\else\textdegree\fi{}C with the results ${D}_{\mathrm{II}}=(2.22\ifmmode\pm\else\textpm\fi{}0.57)\mathrm{exp}[\ensuremath{-}(29.53\ifmmode\pm\else\textpm\fi{}0.29)\ifmmode\times\else\texttimes\fi{}\frac{{10}^{3}}{\mathrm{RT}}] \frac{{\mathrm{cm}}^{2}}{sec},$ ${D}_{\ensuremath{\perp}}=(2.00\ifmmode\pm\else\textpm\fi{}0.54)\mathrm{exp}[\ensuremath{-}(29.92\ifmmode\pm\else\textpm\fi{}0.30)\ifmmode\times\else\texttimes\fi{}\frac{{10}^{3}}{\mathrm{RT}}]\frac{{\mathrm{cm}}^{2}}{sec},$ The diffusion of ${\mathrm{Ga}}^{72}$ was measured over the range from about 240\ifmmode^\circ\else\textdegree\fi{}C to 403\ifmmode^\circ\else\textdegree\fi{}C with the results ${D}_{\mathrm{II}}=(0.016\ifmmode\pm\else\textpm\fi{}0.001)\mathrm{exp}[\ensuremath{-}(18.40\ifmmode\pm\else\textpm\fi{}0.06)\ifmmode\times\else\texttimes\fi{}\frac{{10}^{3}}{\mathrm{RT}}]\frac{{\mathrm{cm}}^{2}}{sec},$ ${D}_{\ensuremath{\perp}}=(0.018\ifmmode\pm\else\textpm\fi{}0.001)\mathrm{exp}[\ensuremath{-}(18.15\ifmmode\pm\else\textpm\fi{}0.08)\ifmmode\times\else\texttimes\fi{}\frac{{10}^{3}}{\mathrm{RT}}]\frac{{\mathrm{cm}}^{2}}{sec},$ Copper diffused at a rate faster than gold but slower than silver. As in silver and gold, ${D}_{\mathrm{II}}g{D}_{\ensuremath{\perp}}$; but the anisotropy was much smaller in Cu than in Ag or Au. The measured values of gallium fall somewhat below those for indium and show ${D}_{\ensuremath{\perp}}g{D}_{\mathrm{II}}$. The relative diffusion rates are qualitatively in accord with the predictions of LeClaire's theory of homovalent diffusion. The sign of anisotropy is explained in terms of the electrostatic interaction between the diffusing ion and the vacancy, and the reduced anisotropy of copper diffusion is interpreted as evidence of a size effect.

Experimental results are presented for the diffusion of silver in silver bromide single crystals for temperatures from 140\ifmmode^\circ\else\textdegree\fi{}C to the melting point. The results are obtained by the sectioning technique, using radioactive tracers. The observed diffusion coefficients fall below the values that would normally be expected from the electrical conductivity by a factor of 0.46 to 0.66. The discrepancy is ascribed to the simultaneous occurrence of two types of interstitialcy jumps, with corrections introduced for differing displacements of interstitials and tracer atoms and for correlation effects according to Bardeen and Herring. The resulting mobilities can be represented as $\ensuremath{\mu}={\ensuremath{\mu}}_{0}\mathrm{exp}(\ensuremath{-}\frac{U}{\mathrm{kT}})$ over a wide range of temperatures with ${\ensuremath{\mu}}_{0}=6.2\ifmmode\times\else\texttimes\fi{}{10}^{\ensuremath{-}3} \mathrm{and} 1.57\ifmmode\times\else\texttimes\fi{}{10}^{\ensuremath{-}3}$ ${\mathrm{cm}}^{2}$/volt-sec and $U=0.078 \mathrm{and} 0.225$ ev, respectively, for collinear and non-collinear jumps. These values are in reasonable agreement with theoretical calculations for silver chloride. The diffusion coefficient is observed to fall increasingly below expected values as the temperature approaches the melting point, and as a possible explanation it is surmised that the activation energy for collinear jumps drops gradually towards zero. A comparison is made to similar results which have been reported for diffusion of silver in silver chloride.

The diffusion of mercury has been studied in high-purity zinc as a function of temperature, using the sectioning technique. The diffusion coefficients as measured in single crystals are given by ${D}_{\mathrm{II}}=(0.056\ifmmode\pm\else\textpm\fi{}0.002)\ifmmode\times\else\texttimes\fi{}\mathrm{exp}[\ensuremath{-}\frac{(19700\ifmmode\pm\else\textpm\fi{}48)}{\mathrm{RT}}]$ ${\mathrm{cm}}^{2}$/sec, and ${D}_{\ensuremath{\perp}}=(0.073\ifmmode\pm\else\textpm\fi{}0.006)\mathrm{exp}[\ensuremath{-}\frac{(20180\ifmmode\pm\else\textpm\fi{}94)}{\mathrm{RT}}]$ ${\mathrm{cm}}^{2}$/sec, where ${D}_{\mathrm{II}}$ and ${D}_{\ensuremath{\perp}}$ are the diffusion coefficients parallel and perpendicular to the hexagonal axis, respectively. Compared to the diffusion of cadmium in zinc, where ${D}_{\ensuremath{\perp}}g{D}_{\mathrm{II}}$, mercury diffuses with a slightly smaller activation energy, and with ${D}_{\mathrm{II}}g{D}_{\ensuremath{\perp}}$. The results are discussed on the basis of the basal and nonbasal vacancy mechanisms, and the influence of the ion size of the diffusing atom is examined to explain the diffusion anisotropy.